Cable MSOs Preach Capacity Over Speed
"Think a little bit about speed versus capacity, and maybe we're just barking up the wrong tree when we keep saying 150, 250, 450 [Mbit/s], whatever it is. It's a ridiculous number," said Douglas Semon, vice president of technologies and standards for Time Warner Cable Inc. (NYSE: TWC). "What do our customers really want? We're going to find out. I personally don't think it's about speed wars, I think it's about capacity."
Semon maintains that the FiOS-like idea of delivering 100 Mbit/s to a customer is crazy, because an extremely small percentage of users would have a need for such speed. But by increasing capacity using Docsis 3.0, you have something every customer could appreciate: bandwidth that will not slow down. Consumers' connections might not go faster, but they'll get less of the congestion that slows them down to frustrating levels.
But if a minimal number of customers care about higher speeds, why are a significant number of them fleeing cable for the faster speeds of FTTH telco services like FiOS? (See Verizon Leads the Great 100-Mbit/s Bandwidth Race and Verizon Spells Out 100 Mbit/s.)
"This would be a really good chance to insult all our customers wouldn't it?" said Semon. "Has anyone ever really explained what this means to the end user? It's one of those things where you assume that, oh, faster is better, so I'll switch."
Semon went on to suggest that early on, customers perhaps aren't educated on what a higher speed would actually accomplish, and that they're giving into the marketing claim that faster is by default better. But he maintained that if the economics of the market eventually dictate that speed is king, then Docsis 3.0 can play that game with cable bonding and other tweaks to the network.
The question remains, though, over how much cable will have to dilute its speed through node splits. Cable operators here wouldn’t say how many node splits they have planned for Docsis 3.0, but they did offer alternatives to the problem, such as spectrum jiggling and again, cable bonding.
"We'd like to avoid node splits wherever we can, because they are expensive and they are disruptive to the end user service," Semon said. "I personally believe the driver for a node split shouldn't be just one service like high speed data. It should be holistic. If you need more capacity for any of those services, you should jiggle around the service group in that spectrum, not split the node. But when all of them become congested, then I think it's justifiable to do a node split."
"We're typically dual-fit lasers to nodes that are on an average of 500 homes," offered Joseph Jenson, executive vice president and CTO for Buckeye CableSystem . "Over the last 18 months I'd say downstream has become a real issue for us, and it's becoming an issue of: Do you address that in the spectrum or node splits? It's a difficult situation either way. We do see bonding as being able to father some additional efficiencies."
But as they tweak their networks to bring higher speeds to their customers, the question remains: Do they even know how much speed their customers really want? Some MSOs suggested that they don't, and that the customers don't know either.
"A lot of it is about marketing," said Phil Colby, vice president of technology for Liberty Global Inc. (Nasdaq: LBTY). "It's about customer perception as to what they're getting and what they're paying for it -- and I could probably insult our customers as well. They see the speeds that are being marketed and compare that to the price they're paying, and that's what makes them switch."
John Coppola, director of access technology and engineering at Cox Communications Inc. , agreed: "It's purely marketing perception today."
So while phone companies like Verizon Communications Inc. (NYSE: VZ) say their customers will eventually want unlimited speeds, cable MSOs are saying that they'll never need that bandwidth. But given the fact that customers are giving into the marketing aspect of higher bandwidth, cable is being forced to try to catch up anyway.
— Raymond McConville, Reporter, Light Reading